Squamous cell carcinoma is the most common histology of the oral cavity cancers. There are a multitude of factors that impact the prognosis of patients with these tumors. Amongst these, mandibular bone erosion (through the cortical bone of the mandible: deep cortical and/or marrow) is found to be an important factor (1–5). According to widely accepted staging systems, its presence is considered to be stage T4a (6). The probability of mandibular bone erosion is higher with buccal mucosa lesions in close proximity to the mandible and gingival cancers, which occur due to invasion of the mandible through the occlusal surface (7–9).

Over recent years, it has often been argued that mandibular bone erosion needs to be characterized further. The Japan Society for Oral Tumors (JSOT) has defined T4 cancer as the invasion of the mandibular canal (10–12). Ebrahimi et al. based the T stage on size and depth of invasion for tumor categories T1–T3 and T4 in the presence of marrow invasion (13). In contrast, a few reports have suggested that tumor size correlates with adverse prognosis and that bone invasion is not an independent predictor of survival (14–16). On the contrary, some studies have reported that tumor size and marrow invasion are independent predictors of reduced survival (13, 17–19). In view of such varied evidence and lack of clarity, this study aims to evaluate the association of various patterns of mandibular bone involvement and their impact on survival. Based upon the findings, we also endeavored to develop a staging system that would reflect the implications of various types of bone invasion-superficial cortical erosion (erosive bony involvement), deep cortical erosion (infiltrative bony involvement), marrow involvement (infiltrative bony involvement), and mandibular canal involvement (infiltrative bony involvement), as assessed on imaging in a better way.

This is a retrospective study on treatment-naïve gingivo-buccal complex squamous cell carcinoma (GBC-SCC) patients who underwent preoperative CT imaging between January 1, 2012, and March 31, 2016, at a tertiary care cancer center. The patients who underwent treatment with curative intent were included. Since surgery is the mainstay of treatment for these cancers, only those patients who underwent definitive surgical management at our center were included in the study.

The prognosis of oral squamous cell carcinoma depends upon a multitude of factors. Several of these are included in the staging system. Bone invasion has been considered as an adverse prognostic factor for a long time, thus meriting adjuvant therapy (2). Over the last few decades, it has been shown that superficial cortical erosion for alveolar primaries does not portend a poorer prognosis; such tumors are, therefore, staged according to their size (13).

There have been several studies that have further tried to understand and characterize the type of bone erosion and its effect on prognosis (20, 21). They have differentiated bone erosion as erosive (superficial cortical erosion) and infiltrative (deep cortical, marrow involvement, and mandibular canal involvement) and looked at their impact on the prognosis and survival. It has been observed that cortical bone erosion may not impact survival and only marrow invasion impacts prognosis. A recent meta-analysis found that only marrow invasion impacted overall and disease-free survival (22). On the contrary, few other studies did not show such an association between any type of bone erosion and survival (14–16). Probably this is the reason why staging systems, rather than characterizing the type of bone erosion, continue to mention merely mandibular bone erosion as present or absent.

Studies on this aspect have looked at all subsites of the oral cavity combined. It is prudent to understand that a buccal mucosa or a lower alveolus cancer is more likely to erode mandibular bone as compared with a tongue cancer (8, 9). They cannot be kept on the same pedestal while making any meaningful conclusions regarding upstaging the disease in presence of bone erosion. Another important aspect that these studies have not considered is the pathological depth of invasion, which plays an important role in assessing the prognosis and has recently been incorporated in the AJCC staging system (23, 24). In our study, we utilized the AJCC8 to stage the patients; thus, the depth of invasion was included in the staging process. As mentioned earlier, we only included buccal mucosa and lower alveolus cancer patients in the study, which is the most relevant cohort. We also excluded patients with high infratemporal fossa and maxillary erosion. This was done to exclusively analyze the prognostic impact of the type of mandibular bone erosion on survival.

On multivariate analysis, type of bone erosion had an independent prognostic impact on DFS and LRRFS (p < 0.001 and p < 0.001, respectively) after making adjustments for other prognostic factors ( Tables 3 and 4 , respectively). Deep cortical erosion had survival similar to cases with no bone erosion. In contrast, marrow and mandibular canal involvement had similar survival (DFS and LRRFS), which was statistically worse than that seen with deep cortical erosion and no bone erosion for DFS and LRRFS ( Figures 3 , 4 ).

Based on the results of univariate analysis, patients with deep cortical or no bone involvement were included together, and patients with marrow and mandibular canal were included together for further analysis. We found marrow and mandibular canal involvement to be statistically significantly poorer than no bone or deep cortical erosion for DFS and LRRFS (p =0.023 and p =0.013, respectively). It has also been hypothesized by a few studies that mandibular canal involvement may be associated with higher chances of distant metastasis (25–27). In our study, we found a statistically significant association between type of recurrence and the type of bone erosion (p = 0.013).

There have been few studies that have tried to restage the disease based upon the type of bone erosion. As per Ebrahimi et al., cortex involvement had a similar outcome as no bone involvement (13). They proposed a staging system where the disease was upstaged by 1 T category in the presence of marrow invasion. Another study proposed the JSOT classification, where the tumor was classified as T4a only when there was the involvement of the mandibular canal (10, 11). Involvement of the mandibular marrow without canal involvement was classified according to size; however, these patients performed equally badly as those with canal involvement. Bone erosion was completely ignored in another staging system, where the classification was based upon the soft tissue involvement alone (28). They did not consider bone involvement important for staging the tumor. In all these studies, cases without bone erosion were staged as per the size of the tumor. For staging, they had used the 7th edition of the AJCC, where the impact of depth of invasion was not considered. In the present study, we have staged the patients as per the AJCC8 and have considered the depth of invasion for all the patients.

As per the AJCC8 classification system, the tumor is classified as T4a even on mandibular cortical involvement. But the results of our study show that the cortical involvement did not affect the survival of the patient. Hence, we proposed an MMC classification system in which we downstaged tumors with superficial or deep cortical erosion based solely upon their size and depth of invasion ( Table 1 ). Only those having marrow involvement with or without mandibular canal involvement were staged as T4a. This staging was labeled as MMC. The results of our study also show that T classification based upon the MMC staging was a better predictor of OS, DFS, and LRRFS as compared with the AJCC8 ( Table 5 ).

The limitation of our study is that it is retrospective. We also did not study the impact of superficial bone erosion on the prognosis. Moreover, information about the pattern of invasion on histopathology for these patients was not available. In spite of these limitations, this study provided a large sample size focusing on the relevant subsite of the oral cancers, the buccal mucosa.

In this study, we found that for GBC-SCC, bone erosion with marrow as well as mandibular canal involvement, and not cortical erosion, is associated with poorer survival outcomes. The marrow with or without mandibular canal involvement has a higher incidence of recurrence, and there was a statistically significant association between the type of bone involvement and pattern of recurrence. T classification based upon the proposed Mahajan et al. MMC staging system, which downstages deep cortical bone involvement, is a better predictor for survival as compared with the AJCC8.

The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding author.

The studies involving human participants were reviewed and approved by IEC TMC. The ethics committee waived the requirement of written informed consent for participation.

Study concept: ND, AbM, and AkM. Study design: ND, AbM, and AkM. Data acquisition: ND, AbM, and AS. Quality control of data algorithms: ND, AbM, AkM, and AB. Statistical analysis: ND, AbM, AkM, and AB. Manuscript preparation: all authors. Manuscript editing: ND, AbM, and AkM. Manuscript reviewing: all authors.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.